JP2016152480A - Communication network redundant path detection system and detection method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a communication network redundant path detection system and detection method for the system, that are capable of detecting information flowing repeatedly in the same path on a physical network of a communication system across data centers.SOLUTION: A communication network redundant path detection system S of the present invention comprises: an information collection processing unit s1 for collecting information on start and end points of a logical network of a physical network path, flow information in which a packet processing method for each logical switching means and each physical switching means sw in each data center D is specified, and function information on each server; a network path configuration unit s2 that receives, as input, the collected information, acquires information on a connection situation of each data center D, information on connection between a plurality of data centers Ds, and each server's logical/physical correspondence information, and derives a physical network path 4; a physical redundant path detection unit s3 for detecting a redundant location on the physical network path 4; and a detection result output unit s4.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a redundant route detection system for a communication network in a wide area communication network across data centers and a detection method thereof.

Conventionally, in a communication system, a plurality of logical networks can be conceptually constructed on an actual physical network by network virtualization technology.
Since logical networks are conceptually constructed, they can be constructed without being constrained by the actual physical network structure.

  For example, in the virtual network in the data center, the physical switch has a Spine & Leaf (trunk and leaf) configuration, and the upper switch sw20 and the lower switch sw10 are connected by a full mesh (see FIG. 8). In other words, full mesh connection means that all upper switches sw20 and all lower switches sw10 are connected. Therefore, the network configuration and its communication path are clear.

FIG. 8 is a configuration diagram showing the configuration of the virtual network in the data center.
As shown in FIG. 8, the group of servers s10 is integrated and connected to each ToR (Top of rack) switch sw10 that is Leaf, and each switch sw10 that is Leaf is connected to all the switches sw20 that are Spine. Has been.

  For example, the communication between the servers s10 in the data center shown in FIG. 8 includes the server s10 → ToR switch sw10 → server s10, and the server s10 → ToR switch sw10 → Spine switch sw20 → ToR switch sw10 → server s10. It becomes the route etc.

Suzuki Kazuya et al., "OpenFlow Technology and its Applications", Japan Software Science Society Computer Software, Vol.30 (2013) No.2, P2-13

  On the other hand, in a wide-area virtual network across data centers, the connection between switches / routers between data centers is not a full mesh, so depending on the arrangement of routers in the logical network, packets flow multiple times on the same route on the physical network. A case can occur (see FIG. 9B).

FIG. 9A is a configuration diagram illustrating an example of a logical network, and FIG. 9B is a configuration diagram illustrating an example of a physical network that implements the logical network.
As shown in FIG. 9A, if there is a logical router 103r in the data center 103 due to the configuration of a certain logical network, the route from the server 102s in the data center 102 to a server 104s in the data center 104 is as follows. Via the data center 103.

  By the way, in the actual physical network, for example, as shown in FIG. 9B, the data center 102 has a communication line connected only to the data center 101. The data center 101 is connected to the data center 103 and the data center 104 via a communication line. Each of the data centers 103 and 104 has a communication line connected only to the data center 101. That is, there is no line connecting the data center 102 and the data center 103.

Therefore, on the actual physical network, as shown by the arrow in FIG. 9B, the communication route from the server 102s of the data center 102 to the server 104s of the data center 104 is directly from the server 102s of the data center 102. Since there is no communication line connecting the logical router 103r of the data center 103, the logical router 103r of the data center 103 is reached via the data center 101. Thereafter, the logical router 103r reaches the server 104s of the data center 104 via the data center 101.
As a result, the route connecting the data center 101 and the data center 103 overlaps on the physical network.

Further, when the communication network becomes large in a wide area, the configuration of the communication network becomes complicated. Further, when the above-described case occurs in a situation where a plurality of virtual (logical) networks are configured, it becomes difficult for the operator to grasp the route, which places a heavy burden on the operator.
For example, when a failure occurs in the communication system, it is difficult to identify the failure location. For this reason, the work of identifying the failure location becomes excessive. That is, the maintenance and management work amount increases. In addition, it is difficult to grasp the whole picture of each communication service provided by the communication network.

  The present invention has been devised in view of the above circumstances, and a redundant path detection system for a communication network and a detection method thereof capable of detecting that information flows redundantly on the physical network of a communication system across data centers. For the purpose of provision.

  In order to solve the above-mentioned problem, the redundant route detection system for a communication network according to claim 1 is a redundant route detection system for a communication network that detects a redundant route of a physical network route in a communication network that straddles a plurality of data centers. Information on the start and end points of the logical network of the target physical network path, flow information in which each logical switching means in each data center and the packet processing method of each physical switching means are defined, and each server in the communication network Information collection processing unit that collects the function information, information collected by the information collection processing unit is input, information on the connection status of each data center, connection information between the plurality of data centers, Obtain server logical and physical correspondence information and derive the physical network path Comprising a Ttowaku path components, and the physical redundant path detection unit for detecting a redundant portion that overlaps in the physical network path, and a detection result output unit which outputs the redundant portion.

  According to a seventh aspect of the present invention, there is provided a method for detecting a redundant path detection system for a communication network, comprising an information collection processing unit, a network path configuration unit, a physical redundant path detection unit, and a detection result output unit. A method for detecting a redundant route detection system for a communication network that detects a redundant route of a network route, wherein the information collection processing unit includes information on a start point and an end point of a logical network of the physical network route that is a detection target, Collecting flow information defining the packet processing method of each logical switching means and each physical switching means in the data center, and function information of each server in the communication network, the network path configuration unit is the information collection processing unit The information collected in the above is input, information on the connection status of each data center, A plurality of data center connection information and logical and physical correspondence information of each of the servers, and deriving a physical network path, wherein the physical redundant path detection unit detects an overlapping redundant portion in the physical network path. The detection result output unit outputs the redundant part.

  According to the invention of claim 1 or claim 7, it is possible to detect a redundant route of a physical network route in a communication network across a plurality of data centers. Therefore, the detection of the redundant communication path makes it easy to grasp the communication path and can reduce the burden on the operator.

  The redundant route detection system for a communication network according to claim 2 is the redundant route detection system for a communication network according to claim 1, wherein the flow information includes a condition for identifying a packet sent through the communication network and the condition. Packet processing.

  According to the second aspect of the present invention, the flow information includes a condition for identifying a packet transmitted through the communication network and a packet process that satisfies the condition, so that the packet processing in the switching means can be performed.

  The redundant route detection system for a communication network according to claim 3 is the redundant route detection system for a communication network according to claim 1 or 2, wherein the function information of each server in the communication network is NF FG (Network Function Forwarding). Graph) information.

  According to the third aspect of the present invention, it is possible to generate a route from the start point to the end point via each NF virtual server of the NF FG information to be detected using the NF FG information.

The redundant path detection system for a communication network according to claim 4 is the redundant path detection system for a communication network according to claim 3, wherein the network path configuration unit obtains the NF FG information to be detected from the collected NF FG information. Extracting and generating a route from the start point to the end point via each NF virtual server of the detection target NF FG information based on the start point and end point information of the detection target logical network; From the flow information of the logical switching means, a logical network path corresponding to the path is derived, and from the logical network path, flow information of the physical switching means, connection information between the plurality of data centers, and A physical network path corresponding to the logical network path is derived using logical and physical correspondence information of each server.

  The method for detecting a redundant path detection system for a communication network according to claim 8 is the detection method for a redundant path detection system for a communication network according to claim 7, wherein the information collection processing unit includes function information of each server in the communication network. As NF FG information, and the network path configuration unit extracts the NF FG information of the detection target from the collected NF FG information, and based on the start point and end point information of the logical network of the detection target, Generate a route from the start point to the end point through each NF virtual server of the NF FG information to be detected, and use the flow information of the logic switching means from the route to select a logic corresponding to the route. Deriving a network path, from the logical network path, flow information of the physical switching means, connection information between the plurality of data centers, and the A physical network path corresponding to the logical network path is derived using logical and physical correspondence information of each server.

  According to the invention of claim 4 or claim 8, the physical network path to be detected can be generated.

  The redundant route detection system for a communication network according to claim 5 is the redundant route detection system for a communication network according to any one of claims 1 to 4, wherein the connection status information of each data center includes a logical topology. Information and physical topology information.

  According to the invention of claim 5, the logical topology information can be used to generate a logical network path. Further, the physical topology information can be used for generating a physical network path.

The redundant route detection system for a communication network according to claim 6 is the redundant route detection system for a communication network according to any one of claims 1 to 5, wherein the physical redundant route detection unit is provided in the physical network route. Trace the previous path, store the traced path in the storage unit, and determine whether the destination path is the same as the path traced before being stored in the storage unit. Is detected.

  According to the sixth aspect of the present invention, it is possible to detect a redundant portion of the physical network path to be detected.

  ADVANTAGE OF THE INVENTION According to this invention, the redundant route detection system of the communication network which can detect that the same path | route flows through the same path | route on the physical network of the communication system across a data center, and its detection method can be provided.

The block diagram which shows the hardware constitutions by which the physical redundant path | route detection system of embodiment which concerns on this invention is implement | achieved. The functional block diagram of the physical redundant path | route detection system of embodiment. (A), (b), (c), (d), and (e) are figures which show the specific example of the process of the network path | route structure part of a physical redundant path | route detection system. The figure which shows the detailed structure of the network path | route structure part of a physical redundant path | route detection system. FIG. 3 is a flowchart showing a detailed configuration of a physical redundant path detection unit of the physical redundant path detection system S. (A), (b) is a figure which shows Example 1 and Example 2 of a respectively redundant path | route of a physical network path | route. The figure which shows the outline | summary of the topology detection technique using LLDP. The block diagram which shows the structure of the virtual network in a data center. (A) is a block diagram which shows an example of a logical network, (b) is a block diagram which shows an example of the physical network which implement | achieves the said logical network.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
<< Embodiment 1 >>
FIG. 1 is a configuration diagram showing a hardware configuration in which the physical redundant path detection system according to the embodiment of the present invention is realized.
The physical redundant path detection system S of the embodiment is a system that detects a redundant path of a physical network path to be detected in a communication network.
The physical redundant path detection system S is realized by using software stored in the operation server 1 operated by an operator of the communication network, for example.

The operation server 1 targets a communication network spanning a plurality of data centers D (D1, D2,...).
In each data center D, in the physical network path corresponding to the logical network as shown in FIG. 8, a plurality of servers are integrated in a group of servers, and the physical network path is formed by physical switches sw having the same hierarchical structure as in FIG. Is switched.

  That is, any server in the data center D is connected by switching the physical switch swa (see FIG. 8) for managing each server group and the physical switch swb in the hierarchy above the physical switch. Note that in an actual (physical) data center D, a full mesh connection as shown in FIG. 8 is rare.

  The physical network path is a network path constructed in an actual communication network. On the other hand, the logical network path is a path of the virtual network in accordance with the actual communication network protocol. Hereinafter, “physical” means “actual” in the communication network, and “logic” means “virtual” in the virtual communication network in accordance with the rules of the communication network.

As shown in FIG. 1, each of the data centers D1, D2,... Has an SDN (software defined network) controller c (c1, c2, c3,...).
The SDN controller c1 in the data center D1 controls and manages the SDN-compatible switches sw11, sw12, sw13,... (Sw) in the data center D1.
The SDN controller c2 in the data center D2 controls and manages the SDN-compatible switches sw21, sw22,... (Sw) in the data center D2.

In this manner, each SDN controller c in each data center D controls and manages the SDN-compatible switch sw in each data center D.
That is, all SDN-compatible switches (physical switches) sw of the physical network in the corresponding data center D are managed by the SDN controller c. A logical switch is defined (defined) for the SDN-compatible switch (physical switch) sw.
Each SDN controller c stores the flow information of the physical SDN-compatible switch sw and the flow information of the logical switch.

The operation server 1 is communicatively connected to each SDN controller c1, c2, c3,... In each data center D and the orchestrator o1. That is, the operation server 1 performs input / output of information (signals) between the respective SDN controllers c1, c2, c3,... In each data center D and the orchestrator o1. In other words, the operation server 1 is configured to access the SDN controllers c1, c2, c3,... And the orchestrator o1.
The orchestrator o1 stores and manages FG (Forwarding Graph) information of the network function of the communication network.

In the physical redundant path detection system S, the topology information can be obtained by the topology detection technique using the flow information of the physical network and the logical network in the data center D and LLDP (Link Layer Discovery Protocol).
As described above, the flow information of the physical network and the logical network in the data center D is stored in each SDN controller c.

FIG. 2 is a functional block diagram of the physical redundant path detection system according to the embodiment.
The physical redundant path detection system S includes an information collection processing unit s1, a network path configuration unit s2, a physical redundant path detection unit s3, and a detection result output unit s4.
The information collection processing unit s1, the network path configuration unit s2, the physical redundant path detection unit s3, and the detection result output unit s4 are realized by software described in a program language.

  The information collection processing unit s1 includes information necessary for detecting a physical redundant path (communication service to be detected, start point and end point on the logical network, SDN flow information of each SDN-compatible switch sw, NF FG (Network Function Forwarding Graph) information. And the like are output to the network path configuration unit s2.

  The network path configuration unit s2 acquires topology information, connection information between the data centers D, and logical / physical correspondence information stored in the server. The topology information is information indicating the connection form of the communication network, for example, the form in which the server and the SDN-compatible switch sw are connected.

  The network path configuration unit s2 creates a logical network path using the collected information, and creates a physical network path based on the logical network path. Then, the network path configuration unit s2 outputs the physical network path information to the physical redundant path detection unit s3.

The physical redundant path detection unit s3 detects a physical redundant path from the physical network path and outputs the result to the detection result output unit s4.
The detection result output unit s4 outputs the detected physical redundant path to the outside through a display, a printer, a file output, or the like.
Details of the information collection processing unit s1, the network path configuration unit s2, the physical redundant path detection unit s3, and the detection result output unit s4 will be described later.

<External function of physical redundant path detection system S>
The physical redundant path detection system S has, as external functions, connection information between each SDN controller c, orchestrator o1, data center D, and logical / physical correspondence information stored in each server of the server group in each data center D. Have.

Each SDN controller c of the external function stores flow information (details will be described later) and topology information of an SDN-compatible switch sw (hereinafter referred to as an SDN switch sw) in the corresponding data center D.
The external function orchestrator o1 stores FG information, connection information between the data centers D, and the like.
Logical / physical correspondence information is stored in servers in each data center D.

<Input Information to Information Collection Processing Unit s1>
The information collection processing unit s1 receives as input information information on a communication service (hereinafter referred to as service) provided on the communication network and information on the start point and end point of the service on the logical network. The network manager or the like inputs the service information and the input information of the start point and end point information on the logical network in the operation server 1 (see FIG. 1). Alternatively, the input information is automatically set systematically. In addition, the input form of input information is arbitrary.
For example, a service to be detected is input as input information. For example, X is a service to be detected.

In addition, information on the start point and end point of the service X to be detected on the logical network is input. The start point and the end point are, for example, virtual servers.
Information on the start point and end point on the logical network includes the following information on the start point virtual server and information on the end point virtual server.
{src, dst} = {host_a, host_z}

<Information collection processing section>
In the above logical expression, “src” represents the start point, and “dst” represents the start point. The start virtual server is “a” and the end virtual server is “z”.
Hereinafter, the information collection processing unit s1, the network path configuration unit s2, the physical redundant path detection unit s3, and the detection result output unit s4 of the physical redundant path detection system S will be described in detail.

<Information collection processing unit s1>
The information collection processing unit s1 collects the above-described input information, NF FG information, and SDN flow information.
The information collection processing unit s1 collects (acquires) NF FG information from the orchestrator o1.

<FG information>
The orchestrator o1 holds the FG information.
NF FG is a sequence of NF, for example
FG_1 = (NF_A, NF_B,… NF_C)
It is expressed as here,
NF = {function information, virtual server to be executed}
It is expressed as
“NF_A” indicates that the function information and the virtual server that executes the function information are “A”.

<SDN flow information>
The information collection processing unit s1 collects (acquires) SDN flow information of physical switches and logical switches in each data center D from each SDN controller c.
The SDN flow information is held by the SDN controller c.
The entire flow information can be expressed as follows.
{Flow information of SDN switch 11, flow information of SDN switch sw12, .........} (see FIG. 1)

For example, the flow information column is {Flow1, Flow2, Flow3,.
Note that the SDN switch sw includes a physical switch and a logical switch. A physical switch means a switch provided in an actual communication network, and a logical switch means a switch of a virtual communication network.

The flow information is represented by {header field, action and statistical information}, for example.
The header field is a “condition for identifying a packet”.
The action is “processing of a packet (matching a condition for identifying a packet)”.
This System S basically pays attention to the “Forwarding” action. In this system, “statistical information” is not particularly handled.
The information collection processing unit s1 outputs the collected information to the network path configuration unit s2.

<Network path configuration unit s2>
As illustrated in FIG. 2, the network path configuration unit s2 acquires topology information, connection information between the data centers D, and logical / physical correspondence information.
The network path configuration unit s2 acquires topology information from each SDN controller c. The network path configuration unit s2 acquires connection information between the data centers D from the orchestrator o1, and acquires logical / physical correspondence information from the server.

Connection information between the data centers D is obtained from BGP (Border Gateway Protocol) routing information. BGP is a protocol created for exchanging routes between autonomous systems (AS). The BGP route information includes route information to reach the destination. AS means a set of routers and networks operated with a common policy and the same management.
The server stores logical / physical correspondence information of the server.

First, the network path configuration unit s2 extracts NF FG information corresponding to “service X to be detected” of the input information from the collected NF FG information.
The NF FG information corresponding to service X is as shown in FIG.
{NF_A, NF_B, ........., NF_C} (1)
Suppose that
3A, 3 </ b> B, 3 </ b> C, 3 </ b> D, and 3 </ b> E are diagrams illustrating specific examples of processing of the network path configuration unit of the physical redundant path detection system S. Note that DC1 and DC2 in FIGS. 3D and 3E mean the data center D1 and the data center D2, respectively.

FIG. 4 is a diagram showing a detailed configuration of the network path configuration unit of the physical redundant path detection system S.
The network path configuration unit s2 performs the following process in the logical network path configuration process s1a (see FIG. 4).
In the logical network path configuration processing s1a, a topology detection technique (details will be described later) using LLDP (Link Layer Discovery Protocol) is used.

The logical network path configuration processing s1a extracts the path from the start point (start point server) to the end point (end point server) based on the start point information and end point information of the input information and the NF FG information of the service X. Generated via a virtual server for each NF in the.
For example, as shown in FIG. 3B, a route from the start point (start server) to the end point (end server)
{Vhost_a, vhost_A (NF_A), vhost_B (NF_B),…, vhost_C (NF_C), vhost_z} (2)
And

Here, the prefix “v” means “virtual” (logical). vhost_a, vhost_A, vhost_B,... represent servers a, A, B,. In parentheses are NF (Network Function).
Then, based on “the SDN flow information of the logical switch” and “the logical topology information”, a logical network path between each node (host) is derived.

That is, a logical network path corresponding to the path from the start point (start point server) to the end point (end point server) in (2) is configured.
For example, as shown in FIG.
{Vhost_a, vswA1, vswA2, vhost_A (NF_A), vswB1, vhost_B (NF_B),…, vhost_C (NF_C), vswC2, vhost_z} (3)
It is.
Here, vswA1 and vswA2 are logical switches that constitute a logical network path between vhost_a and vhost_A. Note that a node on the path is a virtual server.

Then, as shown in FIG. 4, the logical network path configuration process s1a outputs the created logical network path information of the service X of (3) to the physical network path configuration process s1b.
Subsequently, the network path configuration unit s2 performs the following processing in the physical network path configuration processing s1b (see FIG. 4).

First, each node (virtual server, virtual switch, virtual router) in the logical network path is converted into a corresponding physical server based on the acquired “logical / physical correspondence information in the server”, and FIG. Deriving a physical server column as shown in FIG.
Here, when each node of the logical network path is converted into a corresponding physical server, the adjacent virtual server and virtual switch may correspond to the same physical server.

Then, a physical network route is derived based on the physical topology information, the physical SDN flow information, and the connection information between the data centers D. That is, the following physical network path (4) (see FIG. 3E) corresponding to the logical network path (3) is generated. When generating a physical network path (see FIG. 3E), a topology detection technique using LLDP is used. The physical network path of service X is, for example, as follows.
{p_host_a, p_swA, p_host_A,… p_swC1, p_host_z} (4)

“P” in the physical network path in (4) means physics. Here, p_swA is a physical node that constitutes a physical network path between p_host_a and p_host_A. In general, a node is a device having an exchange function, a transmission function, a network management function, and the like in an information communication network. A typical node of the Internet is a router, and there are an exchange and a LAN switch. A node on the physical network path is a physical server or a physical switch.

The correspondence between the physical server and the virtual server is based on the logical / physical correspondence information in the server.
As a result, the network path configuration unit s2 outputs the generated physical network path information of the service X (4) to the physical redundant path detection unit s3 (see FIG. 2).

<Physical redundant path detection unit s3>
FIG. 5 is a flowchart showing a detailed configuration of the physical redundant path detection unit of the physical redundant path detection system S.
In the physical redundant path detection unit s3, first, information on the generated physical network path of the service X (see FIG. 3E) is input from the network path configuration unit s2.

Then, a one-hop path from the start point of the physical network path to the next physical node is traced (step s4a in FIG. 5).
Subsequently, a path one hop ahead from the node reached in step s4a to the next physical node in the physical network path is traced (step s4b).

Subsequently, it is determined whether or not the path traced in step s4b is the same path as the previous (or previous) path (step s4c). That is, in step s4c, it is determined whether the route taken is a redundant route.
When it is determined that the path is redundant (Yes in step s4c), information on the redundant physical path is saved (stored) in the storage unit (memory) (step s4d).

On the other hand, when it is determined that the route is not a redundant route (No in step s4c), it is determined whether or not it is the end point of the physical network route (step s4e).
If it is determined that the end point of the physical network path (Yes in step s4e), the process ends.
On the other hand, when it is determined that it is not the end point of the physical network path (No in step s4e), the process proceeds to step s4b. Hereinafter, the same processing is performed.

The redundant path determined in step s4c is the path returning to the previous physical node {a, b, a} (see FIG. 6A) or the path returning to the previous node with a certain length {a, b, c, b, a} (see FIG. 6B).
FIGS. 6A and 6B are diagrams showing Example 1 {..., a, b, a,...} And Example 2 {..., a, b, c, b, a,. It is.

In Example 1 of FIG. 6A, there are a node a and a node b.
In step s4a or step s4b in FIG. 5, after following node b from node a and then following node a from node b in step s4b, it is determined as a redundant path in step s4c.
Thus, in example 1 {..., A, b, a,.

In Example 2 of FIG. 6B, there are a node a, a node b, and a node c. In step s4a or step s4b, after following node b from node a, then following step c from node b to node c, step s4c, step s4b, node c to node b In step s4c, it is determined that the route is redundant.
Further, when the node a is traced from the node b in step s4b via the step s4c, it is determined as a redundant route in step s4c.
In example 2 {..., a, b, c, b, a, ...} of FIG. 6B, a redundant path having a length of 2 is detected.
When the processing of the physical redundant path detection unit s3 is completed in this way, detection result information is output to the detection result output unit s4 (see FIG. 2).

The detection result output unit s4 outputs information on the detected redundant path in the physical network path.
The output form can be output in various forms such as display on a display device (not shown), output from a printer (not shown), and output as electronic data to a file or output to another system.

<Topology detection technology using LLDP>
Next, a topology detection technique using LLDP used in the logical network path configuration processing s1a and the physical network path configuration processing s1b shown in FIG. 4 will be described.
FIG. 7 is a diagram showing an overview of a topology detection technique using LLDP.
Create an LLDP packet with the information on port A of switch A embedded, and send it to switch A using a Packet Out message.
Then, the LLDP packet is output from the designated port X of the switch A and reaches the adjacent switch B.

Subsequently, the switch B sends the received LLDP packet to the controller using a Packet In message. Note that the Packet In message includes information related to the reception port Y of the switch B.
Also, information such as switch A and port X is embedded in the LLDP packet. Therefore, by using these pieces of information, the controller can know the connection relationship between the switches A and B.
By repeating this series of procedures for the SDN switch sw connected to the SDN controller c, the topology of the entire logical / physical network path can be grasped.

In summary, according to the configuration of the physical redundant path detection system S described above, the network path information is configured based on the information of the SDN controller c placed in each data center and the NF FG information of the orchestrator o1. .
Thereby, it is possible to realize a system that detects a physical redundant path that has not been conventionally considered between the data centers D.

Therefore, the detection of the redundant communication path makes it easy to grasp the communication path and can reduce the burden on the operator.
For example, when a failure occurs in the communication system, the location of the failure is easily understood. As a result, the amount of work for identifying the failure location is reduced. That is, the maintenance and management workload is reduced.
In addition, it becomes easy to grasp the whole picture of each communication service provided by the communication network.

  Therefore, it is possible to realize a physical redundant path detection system S of a communication network that can suppress the information from flowing multiple times on the same path on the physical network path of the communication system.

The physical redundant path detection system S described above can be applied to the following cases.
The communication path in SFC (service chaining) is confirmed using the physical redundant path detection system S. SFC is a new set of technologies and processes that can dramatically change how networks are designed and operated.

  In an IaaS (Infrastructure as a Service) service, a user detects a redundant route using a physical redundant route detection system S when a virtual network is constructed, and notifies (recommend). Note that IaaS is a service that provides hardware facilities without having hardware assets.

<< Other Embodiments >>
1. In the above-described embodiment, the information collection processing unit, the network path configuration unit, the physical redundant path detection unit, and the detection result output unit have been described separately. However, each unit may be appropriately integrated or further subdivided. It may be configured. For example, the information collection processing unit and the network path configuration unit may be combined into one functional unit.

2. In the above embodiment, the case where the SDN controller is placed in each data center has been described. However, the SDN controller does not necessarily have to be located in each data center.

3. In the embodiment, the case where the physical redundant path detection system S is realized by software is exemplified, but at least a part of the physical redundant path detection system S may be configured by hardware.

4). The configuration described in the embodiment is an example of the configuration described in the claims, and various specific forms are possible within the scope described in the claims.

(3) Logical network route (4) Physical network route D, D1, D2, ... Data center
host_a Logical network start point S Redundant path detection system s1 Information collection processing unit s2 Network path configuration unit s3 Physical redundant path detection unit s4 Detection result output unit
sw, sw11, sw12, sw13, ... SDN-compatible switch (switching means)
vhost_z Logical network endpoint
X service (detection target)

Claims (8)

A redundant path detection system for a communication network that detects a redundant path of a physical network path in a communication network across a plurality of data centers,
Information on the start and end points of the logical network of the physical network path to be detected; flow information in which each logical switching means in each of the data centers and a packet processing method of each physical switching means are defined; An information collection processing unit for collecting server function information;
Information collected by the information collection processing unit is input, information on the connection status of each data center, connection information between the plurality of data centers, and logical and physical correspondence information of each of the servers, A network path configuration unit for deriving the physical network path;
A physical redundant path detection unit for detecting redundant redundant portions in the physical network path;
A redundant path detection system for a communication network, comprising: a detection result output unit that outputs the redundant portion. The redundant path detection system for a communication network according to claim 1, wherein the flow information includes a condition for identifying a packet transmitted through the communication network and a process for a packet that satisfies the condition. The function information of each server in the communication network is NF FG (Network Function Forwarding Graph) information. The redundant path detection system for a communication network according to claim 1 or 2, wherein: The network path configuration unit includes:
Extract the detected NF FG information from the collected NF FG information,
Based on the information on the start point and end point of the logical network to be detected, generate a route from the start point to the end point through each NF virtual server of the NF FG information to be detected,
From the path, using the flow information of the logic switching means, derive a logical network path corresponding to the path,
From the logical network path, the physical network path corresponding to the logical network path is obtained using the flow information of the physical switching means, the connection information between the plurality of data centers, and the logical and physical correspondence information of each server. The redundant path detection system for a communication network according to claim 3, wherein the system is derived. The redundant path detection system for a communication network according to any one of claims 1 to 4, wherein the information on the connection status of each data center is logical topology information and physical topology information. The physical redundant path detection unit
Trace the previous path in the physical network path, store the traced path in the storage unit, and determine whether the destination path is the same as the path traced before being stored in the storage unit The redundant path detection system for a communication network according to any one of claims 1 to 5, wherein the redundant portion is detected by the detection. An information collection processing unit, a network path configuration unit, a physical redundant path detection unit, and a detection result output unit, and a redundant path detection system for a communication network that detects a redundant path of a physical network path in a communication network across a plurality of data centers. A detection method,
The information collection processing unit is flow information in which the logical network start point and end point information of the physical network path to be detected, and the packet processing method of each logical switching unit and each physical switching unit in each data center are determined. And the function information of each server in the communication network,
The network path configuration unit is input with information collected by the information collection processing unit, information on the connection status of each data center, connection information between the plurality of data centers, and logical and physical information of each server. To obtain the corresponding information, derive the physical network route,
The physical redundant path detection unit detects an overlapping redundant part in the physical network path,
The detection result output unit outputs the redundant part. A detection method of a redundant route detection system of a communication network, characterized in that: The information collection processing unit collects NF FG information as function information of each server in the communication network,
The network path configuration unit includes:
Extract the detected NF FG information from the collected NF FG information,
Based on the information on the start point and end point of the logical network to be detected, generate a route from the start point to the end point through each NF virtual server of the NF FG information to be detected,
From the path, using the flow information of the logic switching means, derive a logical network path corresponding to the path,
From the logical network path, the physical network path corresponding to the logical network path is obtained using the flow information of the physical switching means, the connection information between the plurality of data centers, and the logical and physical correspondence information of each server. The detection method of a redundant route detection system for a communication network according to claim 7, wherein the detection method is derived.

Images